人机分工判据清单
Introduction
Make sure to use this skill whenever the user is deciding how to divide work between a human and an AI, designing a human-in-the-loop / agent loop / evaluator-optimizer workflow, asking "which parts should I let AI do", "how do I verify AI output", "should a human review every step", or planning to delegate a task to an autonomous agent. 也适用于：用户已经写好一套「人类定目标 + AI 循环执行 + 人类验收」方案，要求审查它是否会变成橡皮图章；或用户追问「什么算对」「判据会不会随环境失效」「在开放系统里怎么定对错」「生存下来就是对的能不能当判据」；或用户要用 AI 做科研/编程/研究/内容生产，却说不清完成条件。 它做四件事：① 声明判据的作用域、时间窗、层级代价与目标层复核（G-1）② 前提纠偏（人不是可靠的末端验证者）③ 六道闸门判定任务的可验证性、判定器独立性、能力边界、风险与停止条件 ④ 输出一张可直接执行的「分工判定单」。 Do NOT use for：纯提示词优化（没有分工问题）；纯模型选型/比价；不涉及 AI 的一般项目管理；以及替用户执行对外发布、付款、删除等不可逆动作。
Resource Information
License information is collected from official sources and may be outdated or missing. Refer to the LICENSE file in the original repository before commercial use.
Information boundary
Information on this page is collected from official documents for technical presentation only and does not imply our endorsement. Pricing, quotas and availability are subject to the latest official statements. We make no commitment regarding any action you take after leaving this site.
Similar skills
self-improving agent
Not specified记录自身发现以实现自我改进的技能
Agent Browser
Not specified用于 AI 代理的浏览器自动化 CLI。当用户需要与网站交互（包括浏览页面、填写表单、点击按钮、截图等）时使用。
Memory Setup
API key required启用并配置 Moltbot/Clawdbot 记忆搜索以实现持久上下文。适用于初始化记忆、解决“金鱼脑”问题或协助用户配置 memorySearch。涵盖 MEMORY.md、日志记录及向量搜索设置。
BrowserSkill
Not specifiedBrowserSkill 是腾讯开源的一个智能浏览器自动化工具，让 WorkBuddy 能够像人一样操作网页——自动填表、点击、翻页、截图，完成任何需要浏览器的复杂任务。
Agently Mail
Not specifiedAgently Mail 是 QQ 邮箱团队为 Agent 打造的专属邮箱服务，与个人邮箱隔离，原生适配 Agent，助力你安全、高效地使用 Agent 收发邮件。 Agently Mail 支持微信登录授权、读取邮件、搜索邮件、发送邮件、回复、转发、移动到已删除、上传和下载附件。 Agently Mail 官网 https://agent.qq.com
Agent Memory
Marked paidAPI key required为AI智能体提供持久记忆，用于跨会话存储事实、从行动中学习、回忆信息及追踪实体。
What is a Skill
A Skill is a packaged single capability, such as text recognition, document summarization, or keyword extraction. Developers wrap the calling logic and prompts into a plugin-like unit, so users can load it without understanding the underlying model. StackPilot indexes a large number of skills that can be tried instantly or assembled into a plan.